As I have serial correlation in my residuals up to a very high number of lags, I want to use Newey West standard errors to overcome the serial correlation. Based on the VAR my goal is to conduct forecast error variance decompositions. Does anyone know a way how to replace the standard errors in my VAR object in R by Newey West standard errors? My code is as follow:
var1 <- VAR(Data_Var, lag.max=1, ic="HQ", type = "const")
Neweywest(var1,lag =1, order.by = NULL, prewhite = TRUE, adjust = FALSE,
diagnostics = FALSE, sandwich = TRUE, ar.method = "ols",
data = list(), verbose = FALSE)
I don't know how to combine the VAR with the Newey West standard errors in order to be able to conduct variance decompositions.
Can anyone help me?